Volume 5, Issue 3, March – 2020 International Journal of Innovative Science and Research Technology

ISSN No:-2456-2165

Decentralized Communication Android SDK

1
Rohan Maity 1Dr. M.L. Sharma 1Dr. Krishan Chandra Tripathi
1
Maharaja Agrasen Institute of Technology, Rohini, Delhi, India

Abstract:- This paper introduces Decentralized I. INTRODUCTION

Communications for Chat systems. This is different
from traditional methods to build chat communication  Decentralization Defined
systems such as Whatsapp. Centralized based systems A decentralized network does not work on single or
use centralized server for user sign up, login and user’s central server or machine instead it distributes tasks among
friend list. In this study, an android chat sdk is many nodes/machines.
proposed that is based on decentralized architecture
which uses coroutines for all the networking operations
to perform asynchronous operations without blocking.
The system is controlled by the users. Each user will set
up their own server having the database for friend’s
profiles and each user has to authenticate among each
other before exchanging data or file. SDK will facilitate
developers to create decentralized chat apps for users
with the help of secure and decentralized
communication protocol.

Fig 1:- Centralized System vs Decentralized Systems

The decentralized means there is no central point of  To scale your network, there is the addition of more
control. A lack of a single authority makes the system fairer storage, I/O bandwidth, or processing power to the
and considerably more secure. Since there is no reliance on server. This in turn increase the cost in the long run.
single point there is low risk of data being compromised.
Even if data at point is destroyed would result in no loss of II. METHODOLOGY
data since all the information is stored on multiple devices
around the world. Fundamentals of Decentralization Communication:

Disadvantages of Traditional Centralized Model It includes :

 Decentralized Communication framework (in our case ,
 Risk of a single point of failure . If the central server Matrix protocols)
goes down, the individual “client” machines attached to  Observer Data streams
it are unable to process user requests  Decentralized services
 Limited scalability. All applications and processing
power are dedicated to single server

IJISRT20MAR068 www.ijisrt.com 274

Volume 5, Issue 3, March – 2020 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
 Decentralized Messaging Framework is responsible for Android being the most popular mobile platform, and
message delivery in a peer-to-peer mode. It’s role is to most of the people use chat apps on android, decentralized
manage the observable data streams and they operate on communication android SDK will help developers to create
decentralized services. The Message delivery is based decentralized chat apps for users. Users can use these chat
routers where each router only knows its neighbour apps to communicate without losing power over their data
routers or end-points. to third parties. User will be in control of their own data by
hosting its own server and stay assured of its data not being
 Observable Data streams: these streams have an compromised.
observer attached to it. So that when any changes occur
in the stream, observer is immediately notified. This  Implementation of SDK
concept is taken from Reactive paradigm. We will be using Matrix protocols on which SDK will
work upon. Matrix is a secure and decentralized
 Decentralized Services : these are the microservices communication protocol. This specification is an open
pattern managing server to server connections when standard designed to make privacy and security the default
exchanging data between them. This can be of the web.
implemented by sockets. They tend to be executed as
much as possible in end-users devices. [1] Matrix is an open source and open standard for
decentralised, real-time communication over IP.
 Control of data of third parties in Centralized
architecture:  it's interoperable, meaning it is designed to interoperate
The United States of America. and United Kingdom. with other communication systems, and being an Open
have been forcing WhatsApp, Facebook, and other social Standard means it's easy to see how to interoperate with
media platforms to give the encryption keys of certain it
individuals, allowing these authorities to read their private  Matrix is decentralised, which means there is no central
messages. This treaty has disturbing implications. point - anyone can host their own server and have
WhatsApp uses the Signal Protocol, which guarantees end- control over their data
to-end encryption, preventing even their employees from  it is designed to function in real-time, which means it is
reading your confidential messages. This means the only ideal for building systems that require immediate
way WhatsApp could comply with this is by fundamentally exchange of data, such as Instant Messaging. [4]
altering their software in contradiction to the Signal
protocol. Consider being forced to remove the lock on your Each user connects to a single server, this is their
door to allow the NSA, the FBI, or the police to come visit homeserver. Users are able to participate in rooms that
anytime without notice, warrant, or probable cause. These were created on any Matrix server since each server
demands will render the online communications of federates with other Matrix servers. This means you can
everyone less secure, not just those of suspected criminals. talk to anyone on any server. It also means you can host
your own server, giving you control over all of your data.
 Need of decentralized communication android SDK: Self hosting also gives you the ability to customize your
Decentralization is paramount to restoring privacy and server to fit your needs Each message that is sent in a room
freedom to the internet. As long as there are centralized is synchronized to all of the other servers that participate in
service providers for communication platforms, the that room. If one server goes offline, everyone else in the
government and third parties. will use them to tighten its room can continue talking. Once that server comes back
authoritative grip. online it will be sent all of the messages that it missed
while it was down.

Fig 2:- DCN System architecture

IJISRT20MAR068 www.ijisrt.com 275

Volume 5, Issue 3, March – 2020 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
SDK is built in Kotlin language. very cheap and threads, on the other hand, are resource
heavy to start and manage. A thousand threads can be a
 Coroutines for networking serious challenge for a modern machine.
In function there is no way we can suspend it or
resume it. The only operations on the functions we can All the network calls (like login, authentication,
perform are start and finish. Once the function is started we sending messages etc.) use coroutines for performing non-
must wait until it’s finished. If we call the function again it blocking networking operations
starts execution from it’s beginning.The situation with the
coroutines is different. You can not only start and stop it  Self Hosting of server:
but also suspend and resume it. It is still different than the To host your own server then the reference
kernel’s thread because coroutines are not preemptive by implementation, Synapse could be installed. Synapse is a
themselves (the coroutines on the other hand usually belong reference "homeserver" implementation of Matrix. User
to the thread, which is preemptive). [5] could setup Synapse server, configure the database for
messages, friend list, rooms list etc. own its own.
Kotlin also has great feature “Coroutines”. Coroutines
are a new way of writing asynchronous, non-blocking code. Every user can run one or more clients, which connect
Coroutine are like threads but light weight and like threads, through to a homeserver. The homeserver stores all their
coroutines can run in parallel, wait for each other and personal chat history and user account information.
communicate. The biggest difference is that coroutines are

III. RESULTS

Fig 3:- Initial Set up, users connecting to different server

IJISRT20MAR068 www.ijisrt.com 276

Volume 5, Issue 3, March – 2020 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165

Fig 4:- Select Chat on particular server

Fig 5:- Users connected to different server

IJISRT20MAR068 www.ijisrt.com 277

Volume 5, Issue 3, March – 2020 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165

Fig 6:- User on different Server didn’t receive message

Fig 7:- Users on different servers communicate using matrix protocols

IJISRT20MAR068 www.ijisrt.com 278

Volume 5, Issue 3, March – 2020 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
 Comparison other existing platforms

Factor/ Name Matrix-ios-sdk Matrix-android-sdk Implemented-sdk

Language objective-c/swift java Kotlin

Environment Could be built on MacOS only Could be built on MacOS, Could be built on MacOS,
Linux, Windows Linux, Windows

Set up cost Costlier , since only MacOS is Any machine with 8gb RAM Any machine with 8gb RAM
supported ,i-5 processor works ,i-5 processor works

Asynchronous operation RxSwift, use threads RxJava, uses threads Coroutines, uses light-weight
threads

Code Sharing b/w platforms No, only for iOS No, only for android Yes, with Kotlin/Native code
could be shared between iOS
and android

E2ee, Room, notifications yes yes No

features
Table 1

IV. CONCLUSION

SDK will facilitate developers to create decentralized

chat apps for users with the help of secure and
decentralized communication protocol. And users can have
a self-secured individual user database. User can store all
their personal chat history and user account information
without losing power over their data. User will be in
control of their own data by hosting its own server and stay
assured of its data not being compromised.

REFERENCES

[1]. Paulo Chainho , Steffen Drüsedow†,Ricardo Chaves,

Nuno Santos, Ricardo Lopes Pereira , Haensge† ,
Anton Roman Portabales, Decentralized
Communications: Trustworthy Interoperability in
Peer-To-Peer Networks
[2]. Kundu, Anirban. (2012). Decentralised indexed based
peer to peer chat system. 416-419.
10.1109/ICIEV.2012.6317378.
[3]. Mohamad Afendee Mohamed, Abdullah Muhammed
and Mustafa Man, A Secure Chat Application Based
on Pure Peer-to-Peer Architecture
[4]. M. Patel, B. Naughton, C. Chan, N. Sprecher, S.
Abeta and A. Neal, Decentralised real time
interoperable communication with matrix
[5]. Kahn, Gilles & Macqueen, David. (1977). Coroutines
and Networks of Parallel Processes.
[6]. Kremenova, Iveta & Gajdos, Milan. (2019).
Decentralized Networks: The Future Internet. Mobile
Networks and Applications. 10.1007/s11036-018-
01211-5.

IJISRT20MAR068 www.ijisrt.com 279